Question

我有一个包含四列的Excel电子表格（保存为csv文件）。第一列和第三列包含单词，第二列和第四列包含频率。所以它看起来像这样：

word1, freq1, word2, freq2
word3, freq3, word4, freq4

......等等

我有以下代码将csv转换为json文件。

import csv
import json

csvfile = open('sample.csv', 'r')
jsonfile = open('sample.json', 'w')


fieldnames = ("feature","r", "feature","r")
reader = csv.DictReader(csvfile, fieldnames)
out = json.dumps( [ row for row in reader ] )
jsonfile.write(out)

真的很简单。但是，这会返回一个看起来像这样的json文件

[{"r" : freq2 "feature" : "word2"} {"r" : freq1 "feature" : "word1"}{"r" : freq4 "feature" : "word4"}{"r" : freq3 "feature" : "word3"}]

我正试图找出一种方法，以便返回的json文件看起来像这样

[{"word1" : freq1}{"word2" :freq2}{"word3" :freq3}{"word4" :freq4}]

换句话说，我想将我的电子表格的第一列作为第2列的键，将第3列作为第4列的键。

Answer 1

不幸的是，python的DictReader并不适合你的用例，但是一点zip魔术应该可以解决这个问题。

import csv, json

csvfile = open('sample.csv', 'r')
jsonfile = open('sample.json', 'w')

fieldnames = ("word1","freq1", "word2","freq2")
reader = csv.reader(csvfile, fieldnames)

out = json.dumps( [ dict(zip(row[::2], row[1::2])) for row in reader ] )
jsonfile.write(out)

dict(zip(row[::2], row[1::2]))创建一个字典，从奇数列映射到一行的相应偶数列。

Answer 2

稍微有点罗嗦，但你可以尝试一下吗？

import csv, json

def dump_to_json():                     
    csv_result = []
    with open('sample.csv', 'rb') as csvfile:                
        for row in csv.DictReader(csvfile, delimiter=',', quotechar='"'):
            csv_result.append({'word1': row['word1'], 'freq1': row['freq1'], 
                               'word2': row['word2'], 'freq2': row['freq2']})        

    json_feed = [{c['word1']: c['freq1'], c['word2']: c['freq2']} for c in csv_result]

    with open('sample.json', 'w') as outfile:
        json.dump(json_feed, outfile)       

dump_to_json()

Answer 3

假设你有这样的数据：

feature, r, feature, r
word1, freq1, word2, freq2
word3, freq3, word4, freq4

如果允许我使用自己的库，这是一个说明的解决方案：

>>> import pyexcel
>>> r=pyexcel.SeriesReader("sample.csv")
>>> r[0]
['word1', ' freq1', ' word2', ' freq2']
>>> r[1]
['word3', ' freq3', ' word4', ' freq4']
>>> r.series()
['feature', ' r', ' feature', ' r']
>>> r.column_at(0)
['word1', 'word3']
>>> r.column_at(1)
[' freq1', ' freq3']
>>> r.column_at(2)
[' word2', ' word4']
>>> r.column_at(3)
[' freq2', ' freq4']
>>> a=zip(r.column_at(0),r.column_at(1))
>>> b=zip(r.column_at(2),r.column_at(3))
>>> a+b
[('word1', ' freq1'), ('word3', ' freq3'), (' word2', ' freq2'), (' word4', ' freq4')]
>>> j=open('sample.json', 'w')
>>> import json
>>> j.write(json.dumps(a+b))
>>> j.close()
>>> exit()

结果如下：

[["word1", " freq1"], ["word3", " freq3"], [" word2", " freq2"], [" word4", " freq4"]]

如您所见，引号中还留有空格。所以你可以做的是使用SheetFormatter：

>>> import pyexcel
>>> r=pyexcel.SeriesReader("sample.csv")
>>> def clean(value, type):
...     return value.strip()
...
>>> r.add_formatter(pyexcel.formatters.SheetFormatter(str, clean))
>>> r.column_at(0)
['word1', 'word3']
>>> r.column_at(1)
['freq1', 'freq3']
>>> r.column_at(2)
['word2', 'word4']
>>> r.column_at(3)
['freq2', 'freq4']

更多文档是on pyhosted

Answer 4

尝试使用https://github.com/samarjeet27/CSV-Mapper/

import csvmapper

# create map file
mapper = csvmapper.DictMapper([
     [ 
          {'name':'word1' },
          {'name':'word2'},
          {'name':'word3'},
          {'name':'word4'},
     ]
])
# parser instance
parser = csvmapper.CSVParser('sample.csv', mapper)
converter = csvmapper.JSONConverter(parser)
# conver to json
print converter.doConvert(False)

Answer 5

没有导入的解决方案在PYTHON中将CSV转换为JSON 。

我确定json和csv库运行良好，但我最终没有它们。所以也许这对其他人来说可能很方便。

简而言之：这将从CSV中提取数据并构建JSON字符串。

笨拙，但有效。

#set up paths and vars
csvfile = open('input.csv','r')
jsonfile = open('output.json', 'w')
arr=[]
headers = []

# Read in the headers/first row
for header in csvfile.readline().split(','):
    headers.append(header)

# Extract the information into the "xx" : "yy" format.
for line in csvfile.readlines():  
  lineStr = ''
  for i,item in enumerate(line.split(',')):
    if i < 28:  #I skip the last two columns for my application
        lineStr+='"'+headers[i] +'" : "' + item + '",\n'
  arr.append(lineStr)

csvfile.close()

#convert the array into a JSON string:
jsn = '{\n "entries":['
jsnEnd = ']\n}'
for i in range(len(arr)-1):
    if i == len(arr)-2:
        jsn+="{"+str(arr[i])[:-2]+"}\n" #Get rid of the last comma if last entry
    else:
        jsn+="{"+str(arr[i])[:-2]+"},\n" #Get rid of the last comma
jsn+=jsnEnd

#write to file
jsonfile.write(jsn)
jsonfile.close()
print "Done."

将其更多地添加为其他可能需要这样的脚本的人的参考。

在Python中将CSV转换为Json

5 个答案: